Recent improvements in manufacturing processes and materials properties associated with excellent mechanical characteristics and low weight have became composite materials very attractive for application on civil aircraft structures. However, even new designs are still very conservative, because the composite structure failure phenomena are very complex. Several failure criteria and theories have been developed to describe the damage process and how it evolves, but the solution of the problem is still open. Moreover, modern manufacturing processes, e.g. filament winding, have been used to produce a wide variety of structural shapes. Therefore, this work presents the development of a damage model and its application to simulate the progressive failure of flat composite laminates as well as for composite cylinders made by filament winding process. The proposed damage model has been implemented as a UMAT (User Material Subroutine) and VUMAT (User Material Subroutine for explicit simulations), which were linked to ABAQUSTM Finite Element (FE) commercial package. Progressive failure analyses have been carried out using FE Method in order to simulate the failure of filament wound composite structures under different quasi-static and impact loading conditions. In addition, experiments have been performed not only to identify parameters related to the material model but also to evaluate both the potentialities and the limitations of the proposed model. / As recentes melhorias nos processos de fabricação e nas propriedades dos materiais associadas a excelentes características mecânicas e baixo peso tornam os materiais compósitos muito atrativos para aplicação em estruturas aeronáuticas. No entanto, mesmo novos projetos, ainda são muito conservadores, pois os fenômenos de falha dos compósitos são muito complexos. Vários critérios e teorias de falha têm sido desenvolvidos para descrever o processo de dano e sua evolução, mas a solução do problema ainda está em aberto. Além disso, técnicas modernas de fabricação, como o enrolamento filamentar (filament winding) vêm sendo utilizadas para produzir uma ampla variedade de formas estruturais. Assim, este trabalho apresenta o desenvolvimento de um modelo de dano e a sua aplicação para simular a falha progressiva de estruturas planas e cilíndricas fabricadas em material compósito através do processo de filament winding. O modelo de dano proposto foi implementado como sub-rotinas em linguagem FORTRAN (UMAT-User Material Subroutine e, VUMAT-User Material Subroutine para simulações explícitas), que foram compiladas junto ao programa comercial de Elementos Finitos ABAQUSTM. Várias análises numéricas foram realizadas via elementos finitos, a fim de prever a falha dessas estruturas de material compósito sob diferentes condições de carregamentos quase-estáticos e de impacto. Além disso, vários ensaios experimentais foram realizados, a fim de identificar os parâmetros relacionados com o modelo de material, bem como avaliar as potencialidades e as limitações do modelo proposto.

Les travaux présentés concernent l'exploration des techniques de localisation utilisées en analyse de défaillance dans le but de les appliquer au domaine de la sécurité numérique des circuits et systèmes intégrés. Ces travaux contribuent, d'une part à étendre le champ d'application des techniques d'analyses de vulnérabilités, et d'autre part à apporter des éléments de réponses sur la faiblesse des implémentations cryptographiques sur circuits de type FPGA. Cette thèse s'inscrit donc dans une démarche à la fois de prévention mais aussi de veille technologique en matière d'attaque en apportant un complément d'information sur la faiblesse des implémentations matérielles de systèmes sécurisés. Dans le cadre de l'évaluation des composants de traitement de l'information par les laboratoires agréés (CESTI), l'analyse de vulnérabilité, et plus spécifiquement la cryptanalyse matérielle, a pour but d'éprouver la sécurité des systèmes d'information (composants cryptographiques, carte bancaire, systèmes de cryptage, etc..) dans le but de tester leur résistance face aux attaques connues. En parallèle, dans le cadre de l'analyse de défaillance des circuits utilisés dans le domaine spatial, la localisation de défauts consiste à collecter et analyser les données d'un circuit défaillant afin d'identifier la source du défaut à l'aide de puissants outils. La combinaison de ces deux activités nous a permis dans un premier temps, d'exploiter la lumière émise par un circuit comme un signal de fuite de type « side-channel » par le biais d'une méthode d'attaque semi-invasive par canal auxiliaire, Differential Light Emission Analysis (DLEA). Cette attaque, basée sur un traitement statistique des courbes d'émission de lumière, a permis d'extraire les sous-clés utilisées par un algorithme DES implanté sur circuit FPGA. Dans un second temps, nous avons proposé une seconde technique basée sur la stimulation laser consistant à exploiter l'effet photoélectrique afin d'améliorer les attaques par canaux auxiliaires « classiques ». Pour cela, une attaque DPA améliorée par stimulation laser a été mise en place. Ainsi nous avons démontré que le balayage du faisceau laser sur certains éléments du cryptosystème (algorithme DES implanté sur FPGA) augmente sa signature DPA permettant ainsi de diminuer sensiblement le nombre de courbes de consommation nécessaires pour extraire les sous-clés utilisées par l'algorithme. / The purpose of failure analysis is to locate the source of a defect in order to characterize it, using different techniques (laser stimulation, light emission, electromagnetic emission...). Moreover, the aim of vulnerability analysis, and particularly side-channel analysis, is to observe and collect various leakages information of an integrated circuit (power consumption, electromagnetic emission ...) in order to extract sensitive data. Although these two activities appear to be distincted, they have in common the observation and extraction of information about a circuit behavior. The purpose of this thesis is to explain how and why these activities should be combined. Firstly it is shown that the leakage due to the light emitted during normal operation of a CMOS circuit can be used to set up an attack based on the DPA/DEMA technique. Then a second method based on laser stimulation is presented, improving the “traditional” attacks by injecting a photocurrent, which results in a punctual increase of the power consumption of a circuit. These techniques are demonstrated on an FPGA device.

De nos jours, les mémoires embarquées sont les composants les plus denses dans les "System-On-Chips" (SOCs), représentant actuellement plus que 90% de leur superficie totale. Parmi les différents types de mémoires, les SRAMs sont très largement utilisées dans la conception des SOCs, particulièrement en raison de leur haute performance et haute densité d'intégration. En revanche, les SRAMs conçues en utilisant des technologies submicroniques sont devenus les principaux contributeurs de la consommation d'énergie globale des SOCs. Par conséquent, un effort élevé est actuellement consacré à la conception des SRAMs à faible consommation. En plus, en raison de leur structure dense, les SRAMs sont devenus de plus en plus susceptibles aux défauts physiques comparativement aux autres blocs du circuit, notamment dans les technologies les plus récentes. Par conséquent, les SRAMs se posent actuellement comme le principal détracteur du rendement des SOCs, ce qui cause la nécessité de développer des solutions de test efficaces ciblant ces dispositifs.Dans cette thèse, des simulations électriques ont été réalisées pour prédire les comportements fautifs causés par des défauts réalistes affectant les blocs de circuits spécifiques aux technologies SRAM faible consommation. Selon les comportements fautifs identifiés, différents tests fonctionnels, ainsi que des solutions de tests matériels, ont été proposés pour détecter les défauts étudiés. Par ailleurs, ce travail démontre que les circuits d'écriture et lecture, couramment incorporés dans les SRAMs faible consommation, peuvent être réutilisés pour augmenter le stress dans les SRAMs lors du test, ce qui permet d'améliorer la détection des défauts affectant la mémoire. / Nowadays, embedded memories are the densest components within System-On-Chips (SOCs), accounting for more than 90% of the overall SOC area. Among different types of memories, SRAMs are still widely used for realizing complex SOCs, especially because they allow high access performance, high density and fast integration in CMOS designs. On the other hand, high density SRAMs designed with deep-submicrometer technologies have become the main contributor to the overall SOC power consumption. Hence, there is an increasing need to design low-power SRAMs, which embed mechanisms to reduce their power consumption. Moreover, due to their dense structure, SRAMs are more are more prone to defects compared to other circuit blocks, especially in recent technologies. Hence, SRAMs are arising as the main SOC yield detractor, which raises the need to develop efficient test solutions targeting such devices.In this thesis, failure analysis based on electrical simulations has been exploited to predict faulty behaviors caused by realistic defects affecting circuit blocks that are specific to low-power SRAMs, such as power gating mechanisms and voltage regulation systems. Based on identified faulty behaviors, efficient March tests and low area overhead design for testability schemes have been proposed to detect studied defects. Moreover, the reuse of read and write assist circuits, which are commonly embedded in low-power SRAMs, has been evaluated as an alternative to increase stress in the SRAM during test phase and then improve the defect coverage.

Este trabalho objetivou a caracterização mecânica e a análise de falha de dois tipos de juntas compósitas através de ensaios mecânicos, em que uma das juntas era composta por dois laminados de fibras de carbono recoberto em ambas as faces por tecido de fibra de vidro reforçando uma matriz termoplástica (PPS-C) unidos via soldagem por resistência elétrica, e a outra confeccionada com dois laminados de fibras de carbono reforçando uma matriz de resina epóxi (EPX-C) unidos via colagem por filme de resina epoxídica. Os dois tipos de juntas foram submetidos a impacto único transversal de 10 J, condicionamento higrotérmico, além de carregamento em fadiga compressiva no plano nas mais diversas combinações destes processos degradativos de suas propriedades mecânicas. Observou-se, que a junta termorrígida colada de EPX-C apresentou a maior resistência mecânica em flexão em quatro pontos (F4P) na condição original (como-manufaturada), assim como os maiores valores de resistência residual para as várias condições de degradação mecânica e higrotérmica a que foi submetida. Por sua vez, a junta termoplástica soldada de PPS-C exibiu, em termos percentuais, menores reduções dos valores de resistência à flexão sob as condições avaliadas, relativamente às perdas apresentadas pela junta EPX-C em idênticas circunstâncias. A análise macroscópica da superfície de fratura de ambas as juntas indicou que o modo de falha predominante da junta termorrígida foi interfacial enquanto que, para a junta PPS-C, o modo de falha predominante foi o intralaminar. Análises fratográficas através da microscopia eletrônica de varredura (MEV) evidenciaram para a junta termorrígida EPX-C, uma alta adesão entre fibra/matriz, porém uma relativamente fraca interação entre os aderentes (laminado) e o filme adesivo de colagem, enquanto que, para a junta termoplástica PPS-C, reduzidas interações fibra/matriz forem inferidas na camada externa de PPS-V do aderente assim como entre a malha metálica resistiva e os filmes puros de PPS que a revestiam. Em ensaios de resistência ao cisalhamento interlaminar (ILSS), os espécimes retirados da junta EPX-C na condição virgem evidenciaram uma colagem uniforme/homogênea, enquanto que, para a junta PPS-C, os espécimes de ensaio usinados a partir da junta virgem indicaram a ocorrência de efeitos de degradação térmica altamente localizada nas bordas soldadas. / This study aimed to mechanical characterization and failure analysis of two kinds of composite single-lap joints by mechanical tests, in which a single-lap joint was made of two adherents of carbon fibers coated on both sides with glass fiber fabric reinforcing a thermoplastic matrix (PPS-C) welded via resistance welding, and the another type made of two adherents of carbon fibers reinforcing an epoxy matrix (EPX-C) bonded by epoxy adhesive film. The two types of single-lap joints were subjected to single transverse impact of 10 J, hygrothermal conditioning, and compressive fatigue loading in the plane in various combinations of these degradative processes of mechanical properties. It was observed that the thermosetting bonded joint EPX-C showed the highest strength in four point-flexure test in the original condition (as-manufactured), as well as greater residual strength values for the various conditions of mechanical and hygrothermal degradation which was subjected. In turn, the welded thermoplastic joint PPS-C exhibited, in percentage terms, smaller reductions in flexural strength values under the tested conditions, in respect of losses showed by EPX-C in similar circumstances. Macroscopic analysis of the fracture surfaces from both joints indicated that the predominant failure mode was interfacial to thermosetting while for PPS-C joint, the predominant failure mode was intralaminar. Analysis of fracture surfaces by scanning electron microscopy (SEM) showed, for thermosetting joints EPX-C, a high adhesion between fiber/matrix, but a relatively weak interaction between adherents and the bonding adhesive film, whereas for thermoplastic joint PPS-C, reduced interactions fiber/matrix are inferred in the outer layers of PPS-V and between metal mesh and the pure PPS films that coated it. In the interlaminar shear strength tests (ILSS), the specimens removed from the EPX-C joint in the pristine condition showed a uniform/homogeneous bonding along the joint area, while for PPS-C joint, coupons extracted from pristine condition joint indicated the occurrence of degradation caused by thermal effects localized in the welded edges.

The reliability of protection systems has emerged as an important topic because protection failures have critical influence on the reliability of power systems. The goal of this research is to develop novel approaches for modeling and analysis of the impact of protection system failures on power system reliability. It is shown that repairable and non-repairable assumptions make a remarkable difference in reliability modeling. A typical all-digital protection system architecture is modeled and numerically analyzed. If an all-digital protection system is indeed repairable but is modeled in a non-repairable manner for analysis, the calculated values of reliability indices could be grossly pessimistic. The smart grid is emerging with the penetration of information-age technologies and the development of the Special Protection System (SPS) will be greatly influenced. A conceptual all-digital SPS architecture is proposed for the future smart grid. Calculation of important reliability indices by the network reduction method and the Markov modeling method is illustrated in detail. Two different Markov models are proposed for reliability evaluation of the 2-out-of-3 voting gates structure in a generation rejection scheme. If the model with consideration of both detectable and undetectable logic gate failures is used as a benchmark, the simple model which only considers detectable failures will significantly overestimate the reliability of the 2-out-of-3 voting gates structure. The two types of protection failures, undesired-tripping mode and fail-to-operate mode are discussed. A complete Markov model for current-carrying components is established and its simplified form is then derived. The simplified model can appropriately describe the overall reliability situation of individual components under the circumstances of complex interactions between components due to protection failures. New concepts of the self-down state and the induced-down state are introduced and utilized to build up the composite unit model. Finally, a two-layer Markov model for power systems with protection failures is proposed. It can quantify the impact of protection failures on power system reliability. Using the developed methodology, we can see that the assumption of perfectly reliable protection can introduce errors in reliability evaluation of power systems.

The usage of more inexpensive silicon feedstock for the crystallization of multicrystalline silicon blocks promises cost reduction for the photovoltaic industry. Less expensive substrates made out of metallurgical silicon (MG-Si) are used as a mechanical support for the epitaxial solar cell. Moreover, conventional inert solar cells can be produced from up-graded metallurgical silicon (UMG-Si). This feedstock has higher content of impurities which influences cell performance and mechanical strength of the wafers. Thus, it is of importance to know these effects in order to know which impurities should be preferentially removed or prevented during the crystallization process. Solar cell processing steps can also exert a change in the values of mechanical strength of processed multicrystalline silicon wafers until the fabrication of a solar cell. Bending tests, fracture toughness and dynamic elastic modulus measurements are performed in this work in order to research the mechanical behavior of multicrystalline silicon crystallized with different qualities of silicon feedstock. Bending tests and residual stress measurements allows the quantification of the mechanical strength of the wafers after every solar cell processing step. The experimental results are compared with theoretical models found in the classical literature about the mechanical properties of ceramics. The influence of second phase particles and thermal processes on the mechanical strength of silicon wafers can be predicted and analyzed with the theoretical models. Metals like Al and Cu can decrease the mechanical strength due to micro-cracking of the silicon matrix and introduction of high values of thermal residual stress. Additionally, amorphous silicon oxide particles (SiOx) lower the mechanical strength of multicrystalline silicon due to thermal residual stresses and elastic mismatch with silicon. Silicon nitride particles (Si3N4) reduce fracture toughness and cause failure by radial cracking in its surroundings due to its thermal mismatch with silicon. Finally, silicon carbide (SiC) and crystalline silicon oxide (SiOx) introduce thermal residual stresses but can have a toughening effect on the silicon matrix and hence, increase the mechanical strength of silicon wafers if the particles are smaller than a certain size. The surface of as-cut wafers after multi-wire sawing presents sharp micro-cracks that control their mechanical behavior. Subsequent removal of these micro-cracks by texture or damage etching approximately doubles the mechanical strength of silicon wafers. The mechanical behavior of the wafers is then governed by defects like cracks and particles formed during the crystallization of multicrystalline silicon blocks. Further thermal processing steps have a minor impact on the mechanical strength of the wafers compared to as-cut wafers. Finally, the mechanical strength of final solar cells is comparable to the mechanical strength of as-cut wafers due to the high residual thermal stress introduced after the formation of the metallic contacts which makes silicon prone to crack.

Engineered structures are designed to resist all expected loadings without failure. However, structural failures do occasionally occur due to inadequate design and construction, especially for extreme and abnormal loads. This thesis concerns the progressive collapse of structures due to abnormal loading events, and develops a method of advanced analysis for predicting the progressive collapse behaviour of building structures in the plastic limit state. Combined-stress failure states and stiffness degradation models are proposed to simulate plastic deformation of structural members. Elliptic force-deformation relationships are employed to model the nonlinear material behaviour of members. The stiffness degradation of semirigid connections is modeled by a moment-rotation relationship with four parameters. Having the proposed nonlinear model, a generic member stiffness matrix is derived taking into account elastic-plastic bending, shearing and axial deformations. A computer-based incremental-load nonlinear analysis procedure is developed that progressively updates member stiffness using reduction factors that simulate degraded stiffness behaviour. Three types of localized damage modes are investigated to identify different connection damage scenarios. Account is taken of any debris loading that occurs when disengaged structural components fall onto lower parts of the structure. The associated dynamic effect is taken into account for the quasi-static analysis by utilizing an impact amplification factor. Any progressive collapse occurring thereafter involves a series of failure events associated with topological changes. The progressive-failure analysis procedure is based on the alternate-load-path method suggested in the design and analysis guidelines of the General Services of Administration (GSA, 2003) and the Department of Defense (DoD, 2005). The residual load carrying capacity of the damaged framework is analyzed by incrementally applying prevailing long-term loads and impact debris loads. The deterioration of structural strength is progressively traced to the state at which either global stability is reached or progressive collapse to ground level occurs for part or all of the structure. The analysis procedure is extensively illustrated for several planar steel moment frames, including account for the influence of damaged connections and semi-rigid connection behaviour. The results obtained demonstrate that the proposed method is potentially a powerful tool for the analysis of steel building structures under normal and abnormal loads.

Engineered structures are designed to resist all expected loadings without failure. However, structural failures do occasionally occur due to inadequate design and construction, especially for extreme and abnormal loads. This thesis concerns the progressive collapse of structures due to abnormal loading events, and develops a method of advanced analysis for predicting the progressive collapse behaviour of building structures in the plastic limit state. Combined-stress failure states and stiffness degradation models are proposed to simulate plastic deformation of structural members. Elliptic force-deformation relationships are employed to model the nonlinear material behaviour of members. The stiffness degradation of semirigid connections is modeled by a moment-rotation relationship with four parameters. Having the proposed nonlinear model, a generic member stiffness matrix is derived taking into account elastic-plastic bending, shearing and axial deformations. A computer-based incremental-load nonlinear analysis procedure is developed that progressively updates member stiffness using reduction factors that simulate degraded stiffness behaviour. Three types of localized damage modes are investigated to identify different connection damage scenarios. Account is taken of any debris loading that occurs when disengaged structural components fall onto lower parts of the structure. The associated dynamic effect is taken into account for the quasi-static analysis by utilizing an impact amplification factor. Any progressive collapse occurring thereafter involves a series of failure events associated with topological changes. The progressive-failure analysis procedure is based on the alternate-load-path method suggested in the design and analysis guidelines of the General Services of Administration (GSA, 2003) and the Department of Defense (DoD, 2005). The residual load carrying capacity of the damaged framework is analyzed by incrementally applying prevailing long-term loads and impact debris loads. The deterioration of structural strength is progressively traced to the state at which either global stability is reached or progressive collapse to ground level occurs for part or all of the structure. The analysis procedure is extensively illustrated for several planar steel moment frames, including account for the influence of damaged connections and semi-rigid connection behaviour. The results obtained demonstrate that the proposed method is potentially a powerful tool for the analysis of steel building structures under normal and abnormal loads.

In safety-critical, networked embedded systems, it is important that the way in which a fault(s) in one component of the system can propagate throughout the system to other components is analyzed correctly. Many real-world systems, such as modern aircrafts and automobiles, use large-scale networked embedded systems with complex behavior. In this work, we have developed techniques and a software tool, FauPA, that uses those techniques to automate fault-propagation analysis of large-scale, networked embedded systems such as those used in modern aircraft. This work makes three main contributions. 1. Fault propagation analyses. We developed algorithms for two types of analyses: forward analysis and backward analysis. For backward analysis, we developed two techniques: a naive algorithm and an algorithm that uses Datalog. 2. A system description language. We developed a language that we call Communication System Markup Language (CSML) based on XML. A system can be specified concisely and at a high-level in CSML. 3. A GUI-based display of the system and analysis results. We developed a GUI to visualize the system that is specified in CSML. The GUI also lets the user visualize the results of fault-propagation analyses.

In this work, several beam finite element formulations are proposed for failure analysis of planar reinforced concrete beams and frames under monotonic static loading. The localized failure of material is modeled by the embedded strong discontinuity concept, which enhances standard interpolation of displacement (or rotation) with a discontinuous function, associated with an additional kinematic parameter representing jump in displacement (or rotation). The new parameters are local and are condensed on the element level. One stress resultant and two multi-layer beam finite elements are derived. The stress resultant Euler-Bernoulli beam element has embedded discontinuity in rotation. Bending response of the bulk of the element is described by elasto-plastic stress resultant material model. The cohesive relation between the moment and the rotational jump at the softening hinge is described by rigid-plastic model. Axial response is elastic. In the multi-layer beam finite elements, each layer is treated as a bar, made of either concrete or steel. Regular axial strain in a layer is computed according to Euler-Bernoulli or Timoshenko beam theory. Additional axial strain is produced by embedded discontinuity in axial displacement, introduced individually in each layer. Behavior of concrete bars is described by elastodamage model, while elasto-plasticity model is used for steel bars. The cohesive relation between the stress at the discontinuity and the axial displacement jump is described by rigid-damage softening model in concrete bars and by rigid-plastic softening model in steel bars. Shear response in the Timoshenko element is elastic. Finally, the multi-layer Timoshenko beam finite element is upgraded by including viscosity in the softening model. Computer code implementation is presented in detail for the derived elements. An operator split computational procedure is presented for each formulation. The expressions, required for the local computation of inelastic internal variables and for the global computation of the degrees of freedom, are provided. Performance of the derived elements is illustrated on a set of numerical examples, which show that the multi-layer Euler-Bernoulli beam finite element is not reliable, while the stress-resultant Euler-Bernoulli beam and the multi-layer Timoshenko beam finite elements deliver satisfying results.